smids_3x_deit_small_rms_0001_fold2

This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.1115
• Accuracy: 0.8802

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3676	1.0	225	0.3260	0.8652
0.2682	2.0	450	0.4038	0.8369
0.1969	3.0	675	0.3463	0.8569
0.1563	4.0	900	0.3656	0.8869
0.127	5.0	1125	0.4906	0.8885
0.0496	6.0	1350	0.4366	0.8852
0.0874	7.0	1575	0.6811	0.8735
0.0746	8.0	1800	0.4728	0.9002
0.0301	9.0	2025	0.6425	0.8802
0.0064	10.0	2250	0.6457	0.8852
0.021	11.0	2475	0.6671	0.8752
0.0135	12.0	2700	0.6914	0.8852
0.0087	13.0	2925	0.8348	0.8686
0.0257	14.0	3150	0.6378	0.8769
0.0699	15.0	3375	0.7199	0.8885
0.003	16.0	3600	0.7607	0.8869
0.003	17.0	3825	0.7580	0.8819
0.0003	18.0	4050	0.7463	0.8835
0.0005	19.0	4275	0.6721	0.8852
0.0305	20.0	4500	0.7465	0.8785
0.03	21.0	4725	0.8137	0.8752
0.0098	22.0	4950	0.7797	0.8802
0.0223	23.0	5175	0.8830	0.8735
0.0014	24.0	5400	0.9177	0.8752
0.0318	25.0	5625	1.2159	0.8519
0.0263	26.0	5850	0.9640	0.8669
0.0494	27.0	6075	0.9004	0.8702
0.0002	28.0	6300	1.0163	0.8752
0.0354	29.0	6525	1.0067	0.8752
0.0062	30.0	6750	1.0029	0.8785
0.0239	31.0	6975	0.8464	0.8835
0.0305	32.0	7200	0.8764	0.8752
0.0007	33.0	7425	0.8617	0.8769
0.0	34.0	7650	0.9176	0.8785
0.0	35.0	7875	0.9537	0.8885
0.0028	36.0	8100	0.9078	0.8802
0.0	37.0	8325	0.9401	0.8902
0.0066	38.0	8550	0.9351	0.8802
0.0208	39.0	8775	0.9403	0.8869
0.0	40.0	9000	1.0137	0.8852
0.0103	41.0	9225	1.0628	0.8769
0.0	42.0	9450	0.9758	0.8802
0.0	43.0	9675	1.0037	0.8802
0.0	44.0	9900	1.0404	0.8769
0.0	45.0	10125	1.0618	0.8819
0.0	46.0	10350	1.0847	0.8802
0.0	47.0	10575	1.0984	0.8819
0.0	48.0	10800	1.1045	0.8819
0.0023	49.0	11025	1.1110	0.8802
0.0023	50.0	11250	1.1115	0.8802

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2